WO2018184305A1

WO2018184305A1 - Group search method based on social network, device, server and storage medium

Info

Publication number: WO2018184305A1
Application number: PCT/CN2017/090571
Authority: WO
Inventors: 王健宗; 黄章成; 吴天博; 肖京
Original assignee: 平安科技（深圳）有限公司
Priority date: 2017-04-07
Filing date: 2017-06-28
Publication date: 2018-10-11
Also published as: AU2017268599B2; US10268655B2; TW201837749A; CN107092667B; CN107092667A; TWI652586B; AU2017268599A1; SG11201709902TA; EP3608798A4; JP6608972B2; JP2019513245A; US20180300413A1; EP3608798A1

Abstract

Disclosed is a group search method based on a social network, comprising: receiving a group search request sent by a query terminal, wherein the group search request includes a query user identifier, a group size and a group nuclearity carried therein; retrieving a pre-generated social network graph; performing search expansion, taking as a start point a user node corresponding to the query user identifier in the social network graph, and determining, in each search expansion layer, a user node having the most user nodes connected thereto in the social network graph to be a group member, until the number of group members is equal to the group size, wherein first-level expansion layers being expanded are nodes adjoining a query user, and subsequent expansion layers are user nodes adjoining the group members determined in previous expansion layers; determining whether a nuclearity of a determined group to be searched is not less than the group nuclearity, and if so, using the group to be searched as a query result and feeding the same back to the query terminal for display.

Description

Group search method, device, server and storage medium based on social network

The present application claims priority to Chinese Patent Application No. 200910224884.2, entitled "Community Network-Based Group Search Method and Apparatus", filed on April 7, 2017, the entire contents of which are incorporated by reference. In this application.

Technical field

The present invention relates to the field of computer application technologies, and in particular, to a community network based group method, apparatus, server, and storage medium.

Background technique

Online social networking has become an indispensable part of people's lives and has brought enormous challenges to social network analysis. Group is one of the most important attributes of social networks, and research on grouping based on social networks has gradually become a trend.

In life and work, we often need to find groups from social networks to carry out work, prepare dinners and so on. The traditional group search method first determines the possible combinations of all the nodes in the social network data according to the number of groups, and then filters all the combinations of the search through the other conditions of the group one by one, the calculation amount is very large, and the search efficiency is extremely low.

Summary of the invention

According to various embodiments disclosed herein, a social network based group search method and apparatus, a server, and a storage medium are provided.

A social network based group search method, the method comprising:

Receiving a group search request sent by the querying terminal, where the group search request carries a specified query user identifier, a set group size, and a group verification degree, wherein the group nuclearity defines a group member The minimum number of adjacent members;

Retrieving a pre-generated social network map in response to the group lookup request, wherein the social The network map is generated according to social relationship data in a social networking website, the social network map including a plurality of user nodes and a set of edges for connecting the user nodes;

Performing search expansion based on the user node corresponding to the query user identifier according to the social network map, and determining, in each search expansion layer, a user node that has the largest number of connected user nodes in the social network map as a group a member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User node

Forming the determined group member and the querying user into a group to be searched;

Determining whether the group grading of the group to be searched is not less than the set group grading, and if yes, feeding the group to be searched as a query result to the query terminal for display.

A social network based group search device, the device comprising:

a search request module, configured to receive a group search request sent by the query terminal, where the group search request carries a specified query user identifier, a set group size, and a group check degree, where the group core Degree defines the minimum number of members of a group that are adjacent to other members;

a social network map retrieval module, configured to retrieve a pre-generated social network map in response to the group search request, wherein the social network map is generated according to social relationship data in a social networking website, the social network The figure includes a plurality of user nodes and a set of edges for connecting the user nodes;

a hierarchical search module, configured to perform a search expansion of a set number of layers based on the user node corresponding to the query user identifier according to the social network map, and determine one in each search expansion layer in the social network map. Connecting a user node with the largest number of user nodes as a group member, so that the determined number of the group members meets the set group size;

The user node included in the first-level extension layer in the expansion process is an adjacent node of the query user, and the user node included in the next-level extension layer is the group member determined by the upper-level extension layer. Adjacent user nodes;

a group determining module, configured to form the determined group member and the query user into a group to be searched;

The group feedback module is configured to determine whether the group nuclearity of the to-be-searched group is not less than the set group nuclearity, and if yes, feed the to-be-searched group as a query result to the query terminal display.

A server comprising a memory and a processor, the memory storing computer executable instructions, the instructions being executed by the processor, causing the processor to perform the following steps:

Retrieving a pre-generated social network map in response to the group lookup request, wherein the social network map is generated according to social relationship data in a social networking website, the social network map including a plurality of user nodes and Connecting a set of edges of the user node;

One or more non-volatile readable storage media storing computer-executable instructions, the computer-executable instructions being executed by one or more processors, such that the one or more processors perform the steps of:

Retrieving a pre-generated social network map in response to the group lookup request, wherein the social network map is generated according to social relationship data in a social networking website, the social network map including multiple uses a user node and a set of edges for connecting the user node;

Details of one or more embodiments of the present application are set forth in the accompanying drawings and description below. Other features, objects, and advantages of the invention will be apparent from the description and appended claims.

DRAWINGS

1 is an application environment diagram of a social network-based group search method in an embodiment;

2 is a schematic diagram showing the internal structure of a server in an embodiment;

3 is a flow chart of a community network based group lookup method in an embodiment;

4 is a social network diagram in one embodiment;

5A-5B are schematic diagrams of hierarchical search starting from a query user in an embodiment;

6A-6C are groups to be searched by hierarchical search in one embodiment;

7 is a flowchart of a social network graph node pruning in an embodiment;

8 is a schematic diagram of a social network graph node pruning in an embodiment;

FIG. 9 is a flowchart related to group preference when there are a plurality of groups satisfying the query condition in one embodiment;

10 is a structural block diagram of a social network-based group search apparatus in an embodiment;

11 is a structural block diagram of a social network-based group search apparatus in another embodiment;

FIG. 12 is a structural block diagram of a social network based group search apparatus in still another embodiment.

detailed description

The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

As shown in FIG. 1 , in an embodiment, an application environment diagram of a community network-based group search method is provided, where the application environment map includes a query terminal 110 and a server 120. The query terminal 110 can communicate with the server 120 over a network. The query terminal 110 may be at least one of a smartphone, a tablet, a notebook, and a desktop computer, but is not limited thereto. The server 120 may be an independent physical server or a server cluster composed of a plurality of physical servers. The querying terminal 110 sends a group lookup request to the server 120, specifying the querying user, the group size, and the grouping degree. The server 120 performs hierarchical search and expansion on the data source (social network map) based on the given group query requirements (group size and group verification) of the querying terminal, and each search expansion layer Determining a user node with the largest number of connected user nodes in the social network map as a group member, and ensuring the group size requirement in the determining process. After the group is initially determined, the determined group is verified by the verification. If the verification requirement is met, the group information corresponding to the found group is fed back to the query terminal.

As shown in FIG. 2, in one embodiment, a server 120 is provided that includes a processor coupled via a system bus, a non-volatile storage medium, an internal memory, and a network interface. The non-volatile storage medium of the server 120 stores an operating system, a database, and at least one computer executable instruction. When the computer executable instructions are executed by the processor, the processor can be caused to perform a social network based group lookup method as shown in FIG. The database is used to store data, such as storing social network maps and the like involved in the execution of the social network-based group search method. The processor is used to provide computing and control capabilities to support the operation of the entire server 120. The internal memory in the server provides a cached operating environment for operating systems, databases, and computer executable instructions in a non-volatile storage medium. The network interface is used for communication connection with the query terminal 110. Those skilled in the art can understand The structure of the server shown in FIG. 2 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the server to which the solution of the present application is applied. The specific server may include Show more or fewer parts, or combine some parts, or have different part arrangements.

As shown in FIG. 3, in an embodiment, a social network-based group search method is provided. The method is applicable to the server shown in FIG. 2, and specifically includes the following steps:

Step S202: Receive a group search request sent by the querying terminal, where the group search request carries the specified query user identifier, the set group size, and the group core degree, and the group check degree defines the group member adjacency. The minimum number of other members.

The query interface is displayed in the query terminal, and the query condition can be set in the query interface. Specifically, the query conditions include group size and group verification. The group size refers to the number of group members; the group nuclearity refers to the minimum number of other group members in the adjacent group of each group member in the determined group member.

Explain the query conditions in a specific scenario: Suppose Alice wants to host a dinner event for 10 people (including Alice), in order to avoid too strange between the participants, the dinner is too deserted, Alice proposed The following query requirements: 1) the size of the dinner is 10 (including Alice); 2) each participant knows at least 3 other participants. Requirement 1) is the group size, and requirement 2) is the group verification.

As shown in FIG. 4, a group consisting of 6 group members, each group member in the figure is adjacent to at least 3 other group members, and therefore, the group has a group degree of 3.

The user can set the group size and group verification by inputting controls in the query interface. In addition, you can also specify the query user through the query interface. The query user is a specified group member, that is, a group member that must be included. For example, Alice in the example scenario is the query user.

In one embodiment, the social networking site, such as Weibo, MSN, etc., needs to be set up before specifying the querying user. The server will push the corresponding query user identifier to the terminal control corresponding to the querying user according to the selected social networking website. In order to enable the user to query the selection of the user identification through the terminal control.

By default, the query user is the user ID of the querying end user in the selected database. For example, the selected social networking site is Weibo, and the user identifier of the querying terminal is Litch. By default, the querying user is the identifier of Litch in Weibo, such as Litchsweety.

In one embodiment, the user identity of the end user in the designated social networking site can be automatically queried by the registration identity information (eg, phone number, real name, identity number, etc.) in the registration information.

Step S204: Retrieving a pre-generated social network map in response to the group search request, wherein the social network map is generated according to social relationship data in the social networking website, where the social network map includes a plurality of user nodes and is used to connect the user nodes. Side set.

The social network map is generated based on social relationship data in a pre-designated social networking website. A social relationship in a social networking site may be a relationship in which a friend relationship, a mutual interest, and the like are related to each other. The two elements included in social relationships are the relationship between users and users. The social network map generated according to the social relationship identifies the user through the node, and identifies the relationship between the users through the edge between the nodes. FIG. 4 is a social network diagram.

A social network map generated based on social network-based social relationship data is a very large and complex social network image. According to the traditional practice, filtering out the groups that meet the query conditions will be a major project, which takes more time and larger computing resources. In this embodiment, the purpose of determining the central node by layer by layer is to quickly and accurately achieve the purpose of finding a group that satisfies the query condition, as detailed in the subsequent steps.

Step S206: Perform search expansion based on the user node corresponding to the user identifier according to the social network map, and determine, in each search expansion layer, a user node that has the largest number of connected user nodes in the social network map as a group member, until The determined number of group members is equal to the group size, wherein the user node included in the first level expansion layer in the expansion process is the neighbor node of the query user, and the user node included in the next level expansion layer is determined by the upper level expansion layer. The user node to which the group member is adjacent.

First, the location of the query user identifier is located in the social network graph, and then the user node corresponding to the query user identifier is used as a starting point for hierarchical expansion to form an extended subgraph.

Specifically, the user node included in the first extension layer of the extended sub-picture is a user node adjacent to the user node corresponding to the query user. As shown in the social network diagram in FIG. 5A, assuming that the query user specified by the query terminal is the user node v4, the user node included in the first extension layer is a node adjacent to the user node v4, which are respectively v2, v5, and v6. The extended subgraph in Figure 5B.

After determining the user node in the first extension layer, one user node is determined as a group member among the user nodes included in the first extension layer. Specifically, determine the number of adjacent user nodes in the social network graph. Many nodes are members of the group. The number of neighboring nodes in the social network graph of the v2 user node in the first extension layer is 6; the number of neighboring nodes in the social network graph of the v5 user node is 7; the number of neighboring nodes in the social network graph of the v6 user node is 4 . Therefore, v5 is selected as a group member.

After determining the new group member, the user node corresponding to the new group member is used as the new central node to perform the next level of expansion. That is, v5 is used as a new central node to expand the second extension layer. Similarly, the user node included in the second extension layer is a user node adjacent to v5. It should be noted that the user node that has been determined to be a member of the group is not included in the new extension layer. As shown in FIG. 5B, although v4 is a node adjacent to v5, since v4 is a user node that has been determined to be a member of the group, the node v4 is not included in the second extension layer. The user nodes included in the second extension layer are v1, v2, v3, v6, v7, and v9, respectively. The node with the most adjacent nodes determined according to the second extension layer is v2 (the number of adjacent nodes is six), and therefore, it is determined that the node v2 is a group member.

If the specified group size is 4, then you only need to determine the last group member, that is, expand to the third extension layer. Similarly, the user nodes included in the third extension layer are v1, v8, v3, and v6, respectively. In the social network diagram, the number of user nodes adjacent to v1, v8, v3, and v6 are: 3, 4, 4, and 4, respectively. Therefore, the determined group member can be v8 or v3 or v6.

Step S208: The determined group member and the query user are grouped into a group to be searched.

According to step S206, the determined to-be-searched group is {v4, v5, v2, v6} or {v4, v5, v2, v3} or {v4, v5, v2, v8}.

A group network map of the group to be searched is constructed according to the determined relationship between the group members to be searched and the group members (the association relationship can be determined from the social relationship diagram).

Group network diagrams such as {v4, v5, v2, v6}, {v4, v5, v2, v3}, {v4, v5, v2, v8} are sequentially shown in FIGS. 6A, 6B, and 6C.

Step S210: It is determined whether the group nuclearity of the to-be-searched group is not less than the set group nuclearity, and if yes, the to-be-searched group is fed back to the query terminal as a query result.

Determining the degree of the group to be searched, according to FIG. 6A-6C, the degree of {v4, v5, v2, v6} is 3; the degree of {v4, v5, v2, v3} is 2; The verinity of v4, v5, v2, v8} is 1. If the group verifier specified by the query terminal is 3, {v4, v5, v2, v6} is a group that satisfies the query condition. If the group specified by the query terminal is 2, then {v4, v5, v2, v6} and {v4, v5, v2, v3} A group that meets the query criteria. Push the group that meets the query criteria to the query terminal.

The social network-based group search method in this embodiment performs hierarchical expansion centering on the query user, and the adopted extension only depends on the connection relationship of each node in the social network data. The expansion process is simple and fast, and it is not necessary to filter all the combinations that may be searched one by one, and the query efficiency is high. In addition, the selected group members are members with higher degree of nuclearity, which ensures that the found groups have higher intimacy.

In an embodiment, in step S204: responding to the group search request, retrieving a pre-generated social network map generated according to social relationship data in the social networking website, the social network graph including a plurality of user nodes and After the step of connecting the edge set of the user node, the method further includes: performing pruning on the retrieved social network graph. Then, based on the pruned social network map, step S206 is performed: searching for the set number of layers according to the user node corresponding to the query user identifier according to the social network map, and determining a social layer in each search expansion layer. The user node with the largest number of connected user nodes in the network diagram is used as the group member, so that the determined number of group members meets the set group size; wherein the user node included in the first level expansion layer in the expansion process is a query. The user node is a neighboring node, and the user node included in the next-level extension layer is a user node adjacent to the group member determined by the upper-level extension layer.

Specifically, as shown in FIG. 7, the steps of performing pruning processing on the retrieved social network graph include:

Step S302: Construct a breadth-first search tree with the user node corresponding to the query user identifier as a root node, and the breadth-first search tree sequentially traverses all user nodes in the social network map from the query user identifier.

The breadth-first search starts from a certain vertex v0 of the graph, and after accessing v0, sequentially searches for each of the unvisited neighboring points w1, w2, ... that access v0. Then, sequentially search for each of the adjacent points that have not been accessed by accessing w1, the adjacent points of w2 that have not been accessed, .... That is, starting from v0, from the near to the far, the vertices that are in communication with the v0 path and whose path lengths are 1, 2, ... are accessed in order, until all the vertices in the connected graph are accessed once.

According to the social network diagram in FIG. 8(a), the breadth-first search BFS is constructed with the query user v1 as the root node, and FIG. 8(b) is the query user v1 based on the social network map. Breadth-first search tree.

Step S304: Determine, according to the constructed breadth-first search tree, the shortest society of each node to the querying user. Cross distance.

According to the constructed breadth-first search tree, the social distance of each user node from the root node (query user) can be determined, and the social distance determined by the breadth-first search tree is the shortest social distance. As shown in FIG. 8(b), the shortest social distance of v2, v3, and v5 is 1, the shortest social distance of v4, v6, and v8 is 2, and the shortest social distance of v7 is 3.

Step S306: Calculate the difference between the group size and the shortest social distance of each user node. The user node whose difference is not greater than 1 is a pruning node, and the pruning node and the edge corresponding to the pruning node are removed in the social network graph. set.

Assuming that the specified group size is 4, for the user node v7 with the shortest social distance of 3, the difference between the group size 4 and the shortest social distance 3 is not more than 1, so the user node v7 is a pruning node. User nodes with a shortest social distance greater than 3 also belong to the pruning node.

Step S308: determining whether there is a user node that does not meet the nuclear degree requirement in the user node of the pruned social network map, and if yes, removing the user node that does not meet the verification requirement in the pruned social network map, The social network map is updated after pruning.

Due to the requirement for group nuclearity in the query conditions, the group's nuclearity is essentially the nuclearity of the group members. If the user node does not meet the requirements in the searched social network diagram, if it is proposed as a group member, its nuclearity in the group will be even less satisfactory. Therefore, it is necessary to socialize after pruning. The user node in the network diagram performs a check of the core. If there is a user node in the social network diagram after the pruning that does not meet the verification requirement, the user node is pruned again.

It is assumed that the group verification of the search is 3, and after the node v7 that does not satisfy the condition that the shortest social distance does not satisfy the condition, the verinity of the node v8 is 2, which is smaller than the group ver. Therefore, the pruning is again performed on the basis of the pruned social network map, and the node v8 is removed. Finally, the social network diagram after pruning is shown in Figure 8(c).

The hierarchical expansion based on the pruned social network map will be simpler, and the generated extended subgraph will also be simplified, so that the group query is more efficient.

In an embodiment, as shown in FIG. 9, when the group that satisfies the query condition (group size and group verification) by the hierarchical expansion is multiple, the social network-based group search method further includes:

Step S402: determining, according to the social network map, an edge set between each group member in the to-be-searched group, the root According to the determined edge set, a group network diagram of each group to be queried is constructed by using a group member as a user node.

The group to be searched according to the hierarchical search may determine the member information of the group, and the group information may be determined according to the member information of the group and the social network map (if the pruning process is performed, the social network map after the pruning) The relationship between group members. A group node is used as a user node, and an association relationship between the group members generates an edge set between the user nodes, so that a group network map of the group to be searched is generated.

If the set group verification is 1, then 6A, 6B, and 6C are determined to be found groups {v4, v5, v2, v6}, {v4, v5, v2, v3}, and {v4, respectively. Group network diagram for v5, v2, v8}.

Step S404: Calculate the group intimacy of each group network map.

Calculating the group intimacy corresponding to each group to be searched according to the generated group network map. First, calculate the intimacy between any two nodes in the group network diagram. The node intimacy of all possible combinations is then summed to obtain group intimacy.

As shown in FIG. 6A, the node intimacy of {v2, v6}, {v2, v4}, {v2, v5}, {v4, v6}, {v4, v5}, {v5, v6} is sequentially calculated, and then These 6 nodes are intimately summed to obtain a group group density.

Specifically, the formula for calculating the node intimacy is:

Given a network graph G = (V, E), nodes u, v ∈ V and (u, v) ∈ E, then the intimacy between nodes u, v is:

Where N(u) represents the set of adjacent nodes of node u, and the number of adjacent nodes of nodes u and v is |N(u)∩N(v)|. In the calculation of intimacy, the numerator adopts the co-neighbor of the node plus 1, mainly to avoid the case where there is an edge between the two nodes, but the intimacy is zero.

The intimacy of a network graph G = (V, E) is the sum of the intimacy of the edges between its nodes, denoted as Co(G) = ∑ _{(u, v) ∈ E} ω (u, v).

For example, the intimacy of the social network diagram shown in FIG. 6A is Co(G _A )=ω(v ₂ , v ₄ )+ω(v ₂ , v ₅ )+ω(v ₂ ,v ₆ )+ω( v ₄ ,v ₅ )+ω(v ₄ ,v ₆ )+ω(v ₅ ,v ₆ )=3/4+3/4+3/4+3/4+3/4+3/4=9 /2; Co(G _B )=ω(v ₂ , v ₃ )+ω(v ₂ , v ₅ )+ω(v ₂ , v ₄ )+ω(v ₄ , v ₅ )+ω(v ₄ , v ₃ )+ω(v ₅ ,v ₃ )=1/2+3/4+1/2+1/2+3/2+1/2=17/4 Co(G _C )=ω(v ₂ , v ₄ )+ω(v ₂ ,v ₅ )+ω(v ₂ ,v ₈ )+ω(v ₄ ,v ₅ )+ω(v ₄ ,v ₈ )+ω(v ₅ ,v ₈ )= 1/2+1/2+1/4+2/3+1+1=47/12;

By comparison, it is known that Co(G _A )>Co(G _B )>Co(G _C ). Therefore, the group pushed to the inquiry terminal is the group shown in FIG. 6A.

Step S406: The group to be searched corresponding to the group network map with the highest intimacy is fed back to the query terminal as a query result.

In one embodiment, the query result fed back to the query terminal is a group network map generated according to the association relationship between the group members and the group members in the social network map, as shown in FIG. 6A.

In an embodiment, when the number of groups satisfying the query condition (group size and group verification) searched by the hierarchical expansion is plural, the verification degree of each determined group to be searched may also be calculated, if present, For the maximum degree of verification, the group with the highest degree of verification is fed back to the query terminal as the query result. As shown in Figures 6A-6C, the group verdicts are 3, 2, and 1, respectively. If the group qualification of the query condition is 1, the three groups to be searched satisfy the query condition. At this time, the result is pushed according to the size of the group. That is, the group A to be searched with the highest group nuclearity is pushed to the inquiry terminal for display.

In an embodiment, as shown in FIG. 10, a social network-based group search apparatus is provided, and the apparatus includes:

The search requesting module 510 is configured to receive a group search request sent by the querying terminal, where the group search request carries the specified query user identifier, the set group size, and the group verification degree, wherein the group verification degree Limits the minimum number of group members that are adjacent to other members.

The social network map retrieval module 520 is configured to retrieve a pre-generated social network map according to the group search request, where the social network map is generated according to social relationship data in the social networking website, where the social network map includes a plurality of user nodes and The set of edges used to connect to the user node.

The hierarchical search module 530 is configured to perform a search expansion of the set number of layers according to the user node corresponding to the user identifier according to the social network map, and determine, in each search expansion layer, a maximum number of connected user nodes in the social network map. User node as a group member, so that the determined number of group members meets the set group size;

The user node included in the first-level extension layer in the expansion process is a neighboring node of the query user, and the user node included in the next-level extension layer is a user node adjacent to the group member determined by the upper-level extension layer.

The group determining module 540 is configured to form the determined group member and the query user into a group to be searched.

The group feedback module 550 is configured to determine whether the group verification degree of the to-be-searched group is not less than the set group verification degree, and if yes, feed the to-be-searched group as a query result to the query terminal display. .

In an embodiment, as shown in FIG. 11, the social network-based group search apparatus further includes:

a search tree construction module 610, configured to build a breadth-first search tree with the user node corresponding to the query user identifier as a root node, where the breadth-first search tree sequentially traverses all the social network maps from the query user identifier User node.

The shortest social distance calculation module 620 is configured to determine a shortest social distance from each user node to the root node according to the constructed breadth-first search tree.

The pruning module 630 is configured to calculate a difference between a group size and a shortest social distance of each user node, where the user node with a difference of not more than 1 is a pruning node, and the pruning node and the pruning are removed in the social network diagram. The edge set associated with the node generates a social network map after pruning.

In an embodiment, the pruning module 630 is further configured to: find, in the user node of the social network graph after the pruning, whether there is a user node whose node is less than the group nuclearity, wherein the node core The degree is the number of nodes adjacent to the user node, and if so, the found user node is removed from the pruned social network map, and the pruned social network map is updated.

In an embodiment, as shown in FIG. 12, the determined group to be searched is multiple; the social network-based group search device further includes:

a group network diagram construction module 710, configured to determine, according to the social network map, an edge set between each group member in the to-be-searched group, and use the group member as a user node according to the determined edge set Constructing a group network diagram of each of the groups to be queried;

The intimacy calculation module 720 is configured to calculate the intimacy of each group network map;

The group feedback module 550 is further configured to feed back the to-be-searched group corresponding to the most intimate group network map as a query result to the query terminal display.

In one embodiment, the group intimacy of the group network map is calculated as:

Co(G)=∑ _(u,v)∈E ω(u,v)

Where Co(G) is the group intimacy, G=(V, E) is the group network map, and V is the section in the group network diagram. The set of points, E is the edge set in the group network diagram,

Node u, v∈V and (u,v)∈E,

Where N(u) represents the set of neighbor nodes of node u, and the number of common neighbor nodes of nodes u and v is |N(u)∩N(v)|.

Each of the above-described network access behavior recognition devices may be implemented in whole or in part by software, hardware, and combinations thereof. The network interface may be an Ethernet card or a wireless network card. The above modules may be embedded in the hardware in the processor or in the memory in the server, or may be stored in the memory in the server, so that the processor calls the corresponding operations of the above modules. The processor can be a central processing unit (CPU), a microprocessor, a microcontroller, or the like.

A person skilled in the art can understand that all or part of the process of implementing the above embodiment method can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium, as in the embodiment of the present invention. The program may be stored in a storage medium of a computer system and executed by at least one processor in the computer system to implement a process comprising an embodiment of the methods as described above. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

The technical features of the above embodiments may be arbitrarily combined. For the sake of brevity of description, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, It is considered to be the range described in this specification.

The above embodiments are merely illustrative of several embodiments of the present invention, and the description thereof is more specific and detailed, but is not to be construed as limiting the scope of the invention. It should be noted that a number of variations and modifications may be made by those skilled in the art without departing from the spirit and scope of the invention. Therefore, the scope of the invention should be determined by the appended claims.

Claims

A social network based group search method includes:

Receiving a group search request sent by the querying terminal, where the group search request carries a specified query user identifier, a set group size, and a group verification degree, wherein the group nuclearity defines a group member The minimum number of adjacent members;

Retrieving a pre-generated social network map in response to the group lookup request, wherein the social network map is generated according to social relationship data in a social networking website, the social network map including a plurality of user nodes and Connecting a set of edges of the user node;

Performing search expansion based on the user node corresponding to the query user identifier according to the social network map, and determining, in each search expansion layer, a user node that has the largest number of connected user nodes in the social network map as a group a member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User node

Forming the determined group member and the querying user into a group to be searched; and

Determining whether the group grading of the group to be searched is not less than the set group grading, and if yes, feeding the group to be searched as a query result to the query terminal for display.
The method of claim 1, wherein the pre-generated social network map is retrieved in response to the group lookup request, wherein the social network map is based on social relationship data in a social networking website After the social network map includes a plurality of user nodes and a set of edges for connecting the user nodes, the method further includes:

Constructing a breadth-first search tree with the user node corresponding to the query user identifier as a root node, and the breadth-first search tree sequentially traverses all the user nodes in the social network map from the query user identifier;

Determining a shortest social distance of each user node to the root node according to the constructed breadth-first search tree;

Calculating a difference between the group size and a shortest social distance of each of the user nodes, the difference The user node that is not greater than 1 is a pruning node, and the pruning node and the edge set associated with the pruning node are removed in the social network map to generate a pruned social network graph.
The method according to claim 2, wherein in calculating the difference between the group size and the shortest social distance of each of the user nodes, the user node whose difference is not greater than 1 is pruning After the prune node and the edge group associated with the prune node are removed from the social network map to generate the prune social network map, the method further includes:

Finding, in the user node of the social network graph after the pruning, whether there is a user node whose node nuclear degree is smaller than the group nuclear degree, wherein the node nuclearity is the number of nodes adjacent to the user node, and if so, Then, the found user node is removed in the pruned social network map, and the pruned social network map is updated.
The method according to claim 1, wherein the searching for a user base node corresponding to the query user identifier is performed according to the social network map, and each of the search expansion layers determines one of the a user node in the social network graph that has the largest number of connected user nodes as a group member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User nodes include:

Positioning the user node where the query user identifier is located in the social network map, and performing hierarchical expansion with the located user node as a starting point, and generating a first expansion layer of the extended sub-picture, where the first expansion layer The included user node is a user node adjacent to the query user;

Determining a group member from the first extension layer of the extended sub-graph, wherein the user node corresponding to the group member is a user node having the largest number of adjacent nodes in the social network graph;

Performing, by using a user node corresponding to the group member in the first expansion layer as a current starting point, performing an expansion of the next level of the extended sub-picture, and determining a new group member in the corresponding expansion layer, until The determined number of the group members is equal to the group size.
The method according to claim 1, wherein the determined group to be searched is plural; the method further comprises:

Determining, according to the social network map, an edge set between each group member in the to-be-searched group, according to Determining the edge set to construct a group network diagram of each of the to-be-queried groups by using the group member as a user node;

Calculating the intimacy of each of the group network maps;

The to-be-searched group corresponding to the group network map with the highest intimacy is fed back to the query terminal as a query result.
The method according to claim 5, wherein the group intimacy of the group network map is calculated as:

Co(G)=∑ (u,v)∈E ω(u,v)

Co(G) is the group intimacy, G=(V, E) is the group network graph, V is the set of nodes in the group network graph, and E is the edge set in the group network graph.

Node u, v∈V and (u,v)∈E,

Where N(u) represents the set of neighbor nodes of node u, and the number of common neighbor nodes of nodes u and v is |N(u)∩N(v)|.
A group search device based on a social network, comprising:

a search request module, configured to receive a group search request sent by the query terminal, where the group search request carries a specified query user identifier, a set group size, and a group check degree, where the group core Degree defines the minimum number of members of a group that are adjacent to other members;

a social network map retrieval module, configured to retrieve a pre-generated social network map in response to the group search request, wherein the social network map is generated according to social relationship data in a social networking website, the social network The figure includes a plurality of user nodes and a set of edges for connecting the user nodes;

a hierarchical search module, configured to perform a search expansion of a set number of layers based on the user node corresponding to the query user identifier according to the social network map, and determine one in each search expansion layer in the social network map. Connecting a user node with the largest number of user nodes as a group member, so that the determined number of the group members meets the set group size;

Wherein, the user node included in the first level expansion layer in the expansion process is the query user Adjacent node, the user node included in the next-level extension layer is a user node adjacent to the group member determined by the upper-level extension layer;

a group determining module, configured to form the determined group member and the query user into a group to be searched; and

The group feedback module is configured to determine whether the group nuclearity of the to-be-searched group is not less than the set group nuclearity, and if yes, feed the to-be-searched group as a query result to the query terminal display.
The device according to claim 7, further comprising:

a search tree building module, configured to build a breadth-first search tree by using a user node corresponding to the query user identifier as a root node, where the breadth-first search tree sequentially traverses all users in the social network map from the query user identifier node;

a shortest social distance calculation module, configured to determine a shortest social distance of each user node to the root node according to the constructed breadth-first search tree;

a pruning module, configured to calculate a difference between the group size and a shortest social distance of each of the user nodes, where the user node whose difference is not greater than 1 is a pruning node, and is removed in the social network diagram The pruning node and the edge set associated with the pruning node generate a pruned social network map.
The device according to claim 8, wherein the pruning module is further configured to: find, in a user node of the social network graph after the pruning, whether there is a user whose node is less than the group versue a node, wherein the node is a number of nodes adjacent to the user node, and if yes, removing the found user node in the pruned social network map, and updating the pruned social Network Diagram.
The device according to claim 7, wherein the hierarchical search module is further configured to locate the user node where the query user identifier is located in the social network map, and start with the located user node. Performing a hierarchical extension to generate a first extension layer of the extended sub-picture, wherein the user node included in the first extension layer is a user node adjacent to the query user; and determining a group from the first extension layer of the extended sub-picture a group member, wherein the user node corresponding to the group member is a user node having the largest number of adjacent nodes in the social network graph; and the user node corresponding to the group member in the first extension layer is the current starting point Performing expansion of the next level of the expansion subgraph, and determining new group members in the corresponding extension layer until the determined number of the group members is equal to the group Group size.
The device according to claim 7, wherein the determined group to be searched is plural; the device further comprises:

a group network diagram construction module, configured to determine, according to the social network map, an edge set between each group member in the to-be-searched group, and construct the group member as the user node according to the determined edge set a group network diagram of each of the groups to be queried;

a closeness calculation module, configured to calculate an intimacy of each of the group network maps;

The group feedback module is further configured to feed back the to-be-searched group corresponding to the group network map with the highest intimacy as a query result to the query terminal display.
The apparatus according to claim 11, wherein the group intimacy of the group network map is calculated as:

Co(G)=∑ (u,v)∈E ω(u,v)

Co(G) is the group intimacy, G=(V, E) is the group network graph, V is the set of nodes in the group network graph, and E is the edge set in the group network graph.

Node u, v∈V and (u,v)∈E,

Where N(u) represents the set of neighbor nodes of node u, and the number of common neighbor nodes of nodes u and v is |N(u)∩N(v)|.
A server comprising a memory and a processor, the memory storing computer executable instructions, the instructions being executed by the processor, causing the processor to perform the following steps:

Receiving a group search request sent by the querying terminal, where the group search request carries a specified query user identifier, a set group size, and a group verification degree, wherein the group nuclearity defines a group member The minimum number of adjacent members;

Retrieving a pre-generated social network map in response to the group lookup request, wherein the social network map is generated according to social relationship data in a social networking website, the social network map including a plurality of user nodes and Connecting a set of edges of the user node;

Searching according to the social network map with the user node corresponding to the query user identifier as a starting point Extending, in each search expansion layer, determining a user node that has the largest number of connected user nodes in the social network map as a group member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User node

Forming the determined group member and the querying user into a group to be searched; and

Determining whether the group grading of the group to be searched is not less than the set group grading, and if yes, feeding the group to be searched as a query result to the query terminal for display.
The server according to claim 13, wherein the pre-generated social network map is retrieved in response to the group search request, wherein the social network map is based on a social relationship in a social networking website After the data is generated, the social network map includes a plurality of user nodes and a set of edges for connecting the user nodes, and the processor further performs the following steps:

Constructing a breadth-first search tree with the user node corresponding to the query user identifier as a root node, and the breadth-first search tree sequentially traverses all the user nodes in the social network map from the query user identifier;

Determining a shortest social distance of each user node to the root node according to the constructed breadth-first search tree;

Calculating a difference between the group size and a shortest social distance of each of the user nodes, where the user node whose difference is not greater than 1 is a pruning node, and the pruning node is removed in the social network map and A set of edges associated with the pruning node generates a social network map after pruning.
The server according to claim 14, wherein the calculating a difference between the group size and a shortest social distance of each of the user nodes is performed, and the user node whose difference is not greater than 1 is a cut. a branch node, after removing the pruning node and the edge set associated with the pruning node in the social network map, and after generating the pruned social network map, the processor further performs the following steps:

Finding, in the user node of the social network graph after the pruning, whether there is a user node whose node nuclear degree is smaller than the group nuclear degree, wherein the node nuclearity is the number of nodes adjacent to the user node, and if so, Removing the found user node in the pruned social network map and updating the pruning The social network map.
The server according to claim 13, wherein the performing, by the processor, the search and expansion according to the user node corresponding to the query user identifier according to the social network map, and expanding in each search Determining, in the layer, a user node that has the largest number of connected user nodes in the social network graph as a group member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User nodes include:

Positioning the user node where the query user identifier is located in the social network map, and performing hierarchical expansion with the located user node as a starting point, and generating a first expansion layer of the extended sub-picture, where the first expansion layer The included user node is a user node adjacent to the query user;

Determining a group member from the first extension layer of the extended sub-graph, wherein the user node corresponding to the group member is a user node having the largest number of adjacent nodes in the social network graph;

Performing, by using a user node corresponding to the group member in the first expansion layer as a current starting point, performing an expansion of the next level of the extended sub-picture, and determining a new group member in the corresponding expansion layer, until The determined number of the group members is equal to the group size.
The server according to claim 13, wherein the determined group to be searched is plural; the processor further performs the following steps:

Determining, according to the social network map, an edge set between each group member in the to-be-searched group, and constructing, according to the determined edge set, the group member as a user node Group network diagram;

Calculating the intimacy of each of the group network maps;

The to-be-searched group corresponding to the group network map with the highest intimacy is fed back to the query terminal as a query result.
The server according to claim 17, wherein the group intimacy of the group network map is calculated as:

Co(G)=∑ (u,v)∈E ω(u,v)

Co(G) is the group intimacy, G=(V, E) is the group network graph, V is the set of nodes in the group network graph, and E is the edge set in the group network graph.

Node u, v∈V and (u,v)∈E,

Where N(u) represents the set of neighbor nodes of node u, and the number of common neighbor nodes of nodes u and v is |N(u)∩N(v)|.
One or more non-volatile readable storage media storing computer-executable instructions, the computer-executable instructions being executed by one or more processors, such that the one or more processors perform the steps of:

Receiving a group search request sent by the querying terminal, where the group search request carries a specified query user identifier, a set group size, and a group verification degree, wherein the group nuclearity defines a group member The minimum number of adjacent members;

Retrieving a pre-generated social network map in response to the group lookup request, wherein the social network map is generated according to social relationship data in a social networking website, the social network map including a plurality of user nodes and Connecting a set of edges of the user node;

Performing search expansion based on the user node corresponding to the query user identifier according to the social network map, and determining, in each search expansion layer, a user node that has the largest number of connected user nodes in the social network map as a group a member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User node

Forming the determined group member and the querying user into a group to be searched; and

Determining whether the group grading of the group to be searched is not less than the set group grading, and if yes, feeding the group to be searched as a query result to the query terminal for display.
A non-volatile readable storage medium according to claim 19, wherein Retrieving a pre-generated social network map in response to the group lookup request, wherein the social network map is generated according to social relationship data in a social networking website, the social network map including a plurality of user nodes and After connecting the edge set of the user node, the processor also performs the following steps:

Constructing a breadth-first search tree with the user node corresponding to the query user identifier as a root node, and the breadth-first search tree sequentially traverses all the user nodes in the social network map from the query user identifier;

Determining a shortest social distance of each user node to the root node according to the constructed breadth-first search tree;

Calculating a difference between the group size and a shortest social distance of each of the user nodes, where the user node whose difference is not greater than 1 is a pruning node, and the pruning node is removed in the social network map and A set of edges associated with the pruning node generates a social network map after pruning.
The non-volatile readable storage medium according to claim 20, wherein said calculating a difference between said group size and a shortest social distance of each of said user nodes is performed, said difference is not The user node greater than 1 is a pruning node, and after removing the pruning node and the edge set associated with the pruning node in the social network map to generate a pruned social network map, the processor further Perform the following steps:

Finding, in the user node of the social network graph after the pruning, whether there is a user node whose node nuclear degree is smaller than the group nuclear degree, wherein the node nuclearity is the number of nodes adjacent to the user node, and if so, Then, the found user node is removed in the pruned social network map, and the pruned social network map is updated.
The non-volatile readable storage medium according to claim 19, wherein said searching performed by said processor searches for a user node corresponding to said query user identifier according to said social network map Expanding, determining, in each search expansion layer, a user node that has the largest number of connected user nodes in the social network graph as a group member until the determined number of the group members is equal to the group size;

The user node that is included in the extended first-level extension layer is an adjacent node of the query user, and the user node included in the next-level extension layer is adjacent to the group member determined by the upper-level extension layer. User nodes include:

Positioning the user node where the query user identifier is located in the social network map, and performing hierarchical expansion with the located user node as a starting point, and generating a first expansion layer of the extended sub-picture, where the first expansion layer The included user node is a user node adjacent to the query user;

Determining a group member from the first extension layer of the extended sub-graph, wherein the user node corresponding to the group member is a user node having the largest number of adjacent nodes in the social network graph;

Performing, by using a user node corresponding to the group member in the first expansion layer as a current starting point, performing an expansion of the next level of the extended sub-picture, and determining a new group member in the corresponding expansion layer, until The determined number of the group members is equal to the group size.
The non-volatile readable storage medium according to claim 19, wherein the determined group to be searched is plural; the processor further performs the following steps:

Determining, according to the social network map, an edge set between each group member in the to-be-searched group, and constructing, according to the determined edge set, the group member as a user node Group network diagram;

Calculating the intimacy of each of the group network maps;

The to-be-searched group corresponding to the group network map with the highest intimacy is fed back to the query terminal as a query result.
The non-volatile readable storage medium according to claim 23, wherein the group intimacy of the group network map is calculated as:

Co(G)=∑ (u,v)∈E ω(u,v)

Co(G) is the group intimacy, G=(V, E) is the group network graph, V is the set of nodes in the group network graph, and E is the edge set in the group network graph.

Node u, v∈V and (u,v)∈E,

Where N(u) represents the set of neighbor nodes of node u, and the number of common neighbor nodes of nodes u and v is |N(u)∩N(v)|.